基于RBF神经网络PID算法的锅炉水温控制与研究  

Research on Boiler Water Temperature Control Based on RBF Neural Network PID Algorithm

作　　者：于钦敏 YU Qinmin(Jilin Electric POWER Co.,Ltd.,Ministry of Thermal Power,Changchun 130022,China)

机构地区：[1]吉林电力股份有限公司火电部,吉林长春130022

出　　处：《传感器世界》2024年第8期11-17,共7页Sensor World

摘　　要：针对过程控制实验中水温控制的PID算法参数难以整定的问题,提出了一种基于径向基函数(Radial Basis Function,RBF)神经网络整定PID的算法,并以仿真与实验相结合的方式应用于水温控制实验中。通过使用阶跃响应曲线法辨识锅炉水温数学模型;设计用RBF神经网络整定PID参数的方法,针对对象模型进行仿真;最后将整定的PID参数应用于水温控制实验验证,并与常规PID方法进行实验比较。结果表明,RBF神经网络整定PID控制器的控制方法具有更好的鲁棒性和自适应性,能取得良好的控制效果。The paper proposes a PID algorithm based on RBF neural network to solve the problem that the PID parameters of water temperature control in process control experiment are difficult to set.The algorithm is applied to water temperature control experiment by simulation and experiment.Firstly,the step response curve method to obtain the mathematics model of the water temperature is used in the paper.Secondly,the RBF neural network tuning PID controller is designed and applied to the water temperature model by simulation.Finally,the tuned PID parameters are put in the water temperature control experiments and experimental comparison with conventional PID method is made.The results indicate that the RBF neural network tuning PID controller has lots of advantages,such as the simple structure,the strong robustness and the better control effect.In a word,the method has good application value.

关 键 词：RBF神经网络整定PID 建模 仿真 水温控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]